4  Creating ADAMs in R

4.1 Introduction to the ADAM data structure

ADAM stands for Analysis Data Model. This model or structure was introduced to record analysis ready information that can be used at the reporting stage. There are certain procedures like imputation of missing values and creation of analysis flags that are not possible at SDTM level, there are done at the ADAM level. Most of the ADAM data sets are based on SDTM datasets, and therefore before creating any ADAM dataset, the corresponding SDTM datasets should be available.

ADAM data structure is further classified into four major classes;

• Subject level data set (ADSL)

The ADSL contains one record per subject regardless of the type of clinical trial design. It contains variables such as subject-level population flags, planned and actual treatment variables, demographic information, randomization factors, and important dates.

• Basic Data Structure datasets (BDS)

A BDS dataset contains one or more records per subject, per analysis parameter, per analysis time point. Analysis time point is conditionally required, depending on the analysis.Other variables maybe needed to describe more completely the structure of the BDS dataset, eg Baseline Type (BASETYPE) is needed when there is more than one definition of baseline for a given analysis parameter(PARAM) in the same dataset.

Other key variables include the value being analyzed (eg AVAL) and the description of the value being analyzed (eg PARAM).

• Occurrences Data Structure (OCCDS)

• ADAM Other data structure